Project Summary Cryptococcal meningitis (CM) is a severe fungal meningoencephalitis that primarily occurs in individuals with advanced HIV/AIDS in resource-limited settings. Despite scale-up of antiretroviral therapy (ART), HIV-associated CM resulted in an estimated 181,100 deaths in 2014 and up to 15% of HIV-related deaths worldwide. In ART-nave patients with CD4 T-cell counts <100 cells/L, asymptomatic cryptococcal antigenemia predicts cryptococcal meningitis and all-cause mortality. The World Health Organization (WHO) recommends cryptococcal antigen (CrAg) screening to detect early infection in ART-nave individuals with CD4<100 cells/L and targeted high-dose fluconazole in asymptomatic CrAg-positive patients followed by ART at 2 weeks or lumbar puncture (LP) to evaluate for CM in patients with symptoms of meningitis. Despite these recommendations, important questions remain on how to optimize CrAg screening strategies. First, a proportion of asymptomatic patients already have sub-clinical meningitis and may receive inferior therapy with fluconazole rather than amphotericin B- based treatment for meningeal disease. Secondly, there is an absence of outcomes data for CrAg screening in ART-experienced patients, who are likely to become a growing population with adoption of HIV ?test-and-treat? strategies with ART increasingly initiated in the absence of CD4 count or CrAg results. In a pragmatic prospective cohort, we will perform laboratory-based CrAg screening of HIV- infected patients with CD4 counts <100 cells/L in Botswana, a country that recently adopted HIV test- and-treat. We will prospectively follow CrAg-positive patients in clinic to screen for meningitis with LP and start patients without meningitis on high-dose fluconazole according to guidelines or refer patients with meningitis for hospitalization. Six-month incident CM and all-cause mortality will be compared between CrAg-positive ART-nave and ART-experienced patients, as well as a CD4-matched CrAg- negative comparator group, and cost-effectiveness evaluated in the context of observed HIV test-and- treat adoption. Secondly, we will validate a novel point-of-care CrAg lateral flow assay (LFA), CryptoPS (Biosynex S.A.), which provides a semi-quantitative CrAg titer instead of binary (positive/negative) result of the commercially available LFA (IMMY, Norman, OK). High CrAg titers have been shown to predict sub-clinical meningitis, and we will determine the sensitivity and specificity of high CryptoPS titer for identifying sub-clinical meningitis as well as predicting 6-month mortality. This study will: 1) provide important insights into CrAg screening utility, cost-effectiveness and outcomes in the era of HIV test- and-treat; and 2) with validation of a semi-quantitative CrAg assay, may lead to important refinements in CrAg testing strategies that incorporate point-of-care CrAg titer results for stratified patient management.